Soldier pile retaining wall deflection 1

[milkshakelake]

Is there a code reference for allowable deflection in a soldier pile retaining wall? It's about 13' tall with 3x8 lagging, holding back sands. The use of the retained soil is sidewalk and backyard. I'm currently calculating 3" to 5" (depending on the steel size), and it seems high to me. But I want to know what ballpark to be in.




I found an older thread discussing this topic, but it didn't truly get resolved: https://www.eng-tips.com/threads/allowable-deflection-in-soldier-pile-ret-walls.103897/

 

[DaveAtkins]

I don't think there are any hard and fast rules. In your case, how much vertical settlement will result from that much lateral movement, and is the settlement acceptable?

 

[milkshakelake]

I'm not sure how to calculate vertical settlement of a soldier pile. I consider lateral earth/surcharge loads. Since they're not loaded axially like gravity piles, I don't consider vertical settlement.

The previous thread also mentioned that there aren't rules about it, and 1" horizontal deflection is kind of a baseline. The thing is, if the 1" limit were true, many cantilever pile installations wouldn't work design-wise. So I'm trying to see if there's some code reference or published material about this.

 

[KootK]

I've seen transportation agency limits on the order of max [2", 1.5%].

Like a lot of things dirt, much seems to rely on the design loads almost never fully materializing.

 

[milkshakelake]

Thanks @KootK , that seems more reasonable than 1".

Agreed about design loads not materializing. Many soldier piles I designed had very large movement if we're going with calculations, but in reality, it was negligible. Part of it is the large surcharge loads that won't happen, like 60 to 300 psf.

 

[Retrograde]

The deflection limit should reflect the specific situation. What I mean by that is that if there is a historic building close to the wall then you probably should aim for the 1" limit. Similar if there are services located in the ground adjacent to the wall.

It it is just a backyard then maybe 3" is OK.

 

[EireChch]

75mm is way too much for a 4m wall.

That should be 50mm max and even that’s excessive

 

[MTNClimber]

For me, 1 inch is reasonable for retaining sensitive areas and 2 inches for less sensitive areas. Anything more would require input from the contractor and owner about potential issues regarding deflection.

Keep in mind that your calculated deflection may not be accurate. The software we use in-house seems to predict high deflection, especially if we're required to use conservative soil values. Often, we have to caveat design submissions that, based on experience, we anticipate will have less deflection than shown in the software's report.

Regarding references, FHWA-RD-75-130 "Lateral Support Systems and Underpinning," dated 1976 by Goldberg, Jaworski, and Gorden, discusses deflection and settlement in Volumes II and III. In Volume III, they provide a summary of:

For Competent Soils (granular soils, very stiff clays, etc.):

"Maximum displacements are typically in the order of 0.25 percent to 0.35 percent of wall height. The lower range is associated with granular soils; the upper range is related to cohesive soils.

Typically, maximum horizontal and vertical displacements are about equal."

For Weaker Soils (soft to medium clays, organic soils, etc.):

"Maximum displacements typically exceed 1 percent of depth of cut for flexible walls such as steel sheet piling. Concrete diaphragm walls dramatically reduce the magnitude of displacements to about 0.25 percent of the depth of cut -- or about the same as those observed for competent soils."

Typically, the maximum vertical displacements exceed maximum horizontal displacement.

When the excavation is underlain by deep deposits of weak soils, the cumulative total of all displacements occurring below the last placed strut level amounts to about 60 percent of the total measured movement."

 

[milkshakelake]

@Retrograde @EireChch I think 3"/75mm might be okay to design for a backyard, because real life deflection will be much less than that. This is based on what MTNClimber and KootK said.


@MTNClimber Thanks a lot for this! I think this is the closest I'm getting to a reference, and I will definitely check that out.

You gave me an idea. I also agree that the software I use (Shoring8) tends to overestimate deflections by a lot. However, I don't have real life monitoring data of soldier pile deflection of what I designed. Over time, I hope to compile some data about how much these soldier piles actually move, and then I'll compare it to program calculations and hand calculations. I don't even have a first set of data to do this analysis, but hopefully it will show up on a project, because I had to specify soldier pile monitoring on a recent project. If this happens, I'll report back and see how much the difference is. I suspect the real life deflection will be about 30% of the calculated one.

However, I don't even know how to calculate the vertical deflection with no axial load, so I'm not sure how that portion will work.

 

[MTNClimber]

The soldier piles shouldn't significantly deflect vertically, but settlement behind the shoring system can result from lateral deflection and soil loss between the lagging.